The Too Big to Ignore explains why Big Data is so important in business and Big Data analytics can be revolutionary for your company. The book takes you step-by-step to learn the tools you need to use Big Data and the future of Big Data for businesses and people's lives.
Who should read this book?
- Those who are interested in business or business
- Those who are interested in technology
- Those who are learning about Big Data to apply to their business activities
About the author
Phil Simon is a technology consultant and author of The Age of the Platform . His work has been featured on such media channels as NBC, CNBC, ABC, as well as BusinessWeek and the Huffington Post magazines .
Why is Big Data important and how to apply the power of Big Data in business activities?
What do you do when you hear people talk about Big Data? Pretend to nod but in my heart I hope no one will ask me about it? Or wondering if I should learn what Big Data really is? Or just treat it as some buzzword?
It's time to think seriously about Big Data. As Phil Simon mentions in this book, Big Data is not simply too big, it is too important to be underestimated.
This book shows why Big Data has become so involved all of a sudden and how you can take full advantage of it. That is also the reason why the book is on the list of books not to be missed. If you want to improve your business or organization, you need to know how to use Big Data.
After reading this book, you will know how to:
– How to align your business towards using Big Data
– How to illustrate Big Data
– Companies like Netflix use Big Data how to promote business activities, grow quickly
Changes in consumption patterns and increasingly cheaper technology have spurred the development of Big Data
Nowadays, everyone is talking about Big Data and many companies are recruiting data experts to understand and apply Big Data in business operations. So what exactly is Big Data? And why is it so influential right now?
Big Data refers to data sets that are too huge and complex to be handled easily. These datasets cannot be processed using only basic software such as Microsoft Excel or Access.
Its growth is driven by the very fact that we are changing our consumption patterns. Smartphones, cloud computing or broadband connections make us constantly consume and create data every minute, every second.
This new, fast and simple way of connecting ensures that we are always “online”, almost anywhere. Think about the first thing you will do when your flight lands?
Turn on your phone and then check your email, Facebook or Twitter. When you do that, you're not just accessing the data, you're creating it. It is these activities that create Big Data.
Our consumption patterns have changed thanks to cheaper technology. Technology costs – especially for data storage and bandwidth – have become incredibly low today. This change is indisputable. Modern technology at affordable prices has allowed and even encouraged us to contribute to Big Data. How many TV series would you download if each episode cost you $10 in bandwidth, not counting the total cost of renting or buying it?
If in 1990, it cost you $10,000 to store 1GB of data, in 2010, this cost dropped to only 10 cents. Reasonable technology costs and Big Data allows uploading to Youtube 48 hours of video per minute. It also allows more than 200 billion videos to be viewed every month. Without Big Data, would these be possible? So what are we going to do with it?
Big Data can give you deeper insights into your customers and business
So what makes Big Data different from other types of data, such as data collected in the 1950s?
It is not only a matter of the volume of data but also the format of the data that has also changed. Most of the data in the Big Data era is unstructured data (a type of data that has not been defined or analyzed before. It can be data in the form of text, information, dates, etc.). In the past, data was often in the form of relational data , this form is simpler and can be easily processed using tables. For example, a table uses one column to display customer information and another column to display product information they have purchased.
In contrast, unstructured data is extremely chaotic. They cannot be arranged in the same way. Each tweet about your product is a piece of data, but you can't simply put tweets in the same table.
Why? Because tweets don't just have a relationship to something else. Imagine if someone tweeted about them reviewing your product. That tweet can contain a lot of information, for example, the person's age, interests, education level, and more. So you simply cannot create a table with “Tweet Line 1” and all the characteristics and information of the tweet in the next column – that would be an unmanageable spreadsheet. floating reason. Now imagine doing the same thing with thousands and thousands of other tweets! It is indeed impossible.
And that's what makes data "unstructured". Today, unstructured data accounts for more than 80% of organizations' data. If we can learn to manage unstructured data, we can understand a lot about consumer behavior. Unstructured data is a huge opportunity indeed.
Netflix (the company that provides movie streaming services) is a perfect example of this. The company tracks information such as where, when and how many times movies are viewed, as well as what device customers use to watch the movies (phone, tablet or personal computer). core…). Netflix also tracks consumer comments about its company on Facebook or Twitter.
In the summer of 2011, Netflix lost about 800,000 customers. When the company looked at customer comments about it on social media, it realized that many customers had abandoned Netflix because the brand and price of their DVD products had been taken over by Qwikster (the service) their courier) reposition. Netflix subsequently terminated the transaction through Qwikster and its business resumed.
Data visualization will allow you to analyze trends
So even if you can organize your unstructured data, how do you analyze and evaluate it? How can something be understood from millions of different pieces of information?
There are two main approaches here. First, time series analysis.
For example, it is easy to predict that sales will increase during Black Friday, but analyzing the data over time will give us more insight into this.
It can show the relationship between sales and the date the customer gets paid – usually the 1st and 15th of every month. It can also distinguish between long-term trends and seasonal trends. An analysis of the data over time can also account for times when sales are volatile – for example, if someone wins the lottery and takes a friend out for a shopping spree.
When you can make sense of your data, you'll be able to ensure that you don't build your strategy on temporary fluctuations. You won't increase your inventory just because a lottery winner visits your store.
The second approach, heat map , helps you to display large amounts of data easily.
Heatmaps represent values in different colors, so can show you more information than conventional methods of displaying data.
A table with 100 million data points probably won't say much; and technically graphs can reveal more information but it also only shows the relationship of two variables. Heatmaps, by contrast, can provide an overview of multiple variables at once. For example, it can evaluate the quantity, content and location of books sold. It also helps your intuition to better understand the data.
You can uncover trends by looking at the color density on the heat map. For example, lots of red dots in an area on the map could indicate high summer sales in a certain area.
Use innovative and groundbreaking platforms to manage Big Data jobs – or outsource.
Here's some news, both good and bad, about Big Data: You'll never use Excel and Access again. You will have to use new technology platforms if you want to succeed with Big Data, and those platforms must be flexible.
Hadoop , for example, is a large collection of data management projects. It does not have a standard configuration. Instead, it is made up of several sub-projects, each of which is quite complex.
The exact jobs of Hadoop are highly technical but it is an essential tool to break down Big Data related jobs into smaller tasks. These subtasks will be processed separately and included in new datasets. Facebook uses Hadoop to analyze huge slices of user data.
If you don't want to spend money on hardware to manage Big Data, you can outsource this to a technology company. If you want, you can test by hiring another company to run Big Data metrics and see if it turns out to be profitable. Kaggle, an online startup created for this purpose.
Kaggle allows you to post Big Data related tasks and find data experts who can help you with those tasks. Even if you don't know what to do with your data, a Kaggle member can suggest you. Members of Kaggle have been given flight and weather information and asked about how to forecast the time of take-off, landing, time to leave the airport or arrive at the airport gate of the plane. They must help schedule flights to adapt to changing conditions. The winner forecast was 40% more accurate than airline standards.
In short, you have to find yourself the best way to manage Big Data. Always question whether you're doing what's best for you.
Make sure your company is really ready for the Big Data era
Even when you're ready for Big Data, you still need a pause. You have to make sure your whole company is as ready as you are.
While there are a number of free tools available, data collection and use will still cost you.
Hadoop is completely free, but you still have to spend a significant part of your budget on consulting and training, to ensure that the investment in Big Data is worth it.
Don't think that Big Data is just a new, profitable and efficient program once you install it. Big Data requires you to restructure your approach to technology and data in general.
Explorys, a company that uses Big Data to improve the quality of medical services realized this when they started. They have to build a network to store data, develop a platform to be able to work with different medical providers. At the same time, they also had to set up a new department with more than 100 employees.
And even if you have a great Big Data tool, remember that it only really helps when you have a good data source. Even the best tools are meaningless so you don't collect useful data.
So start by asking specific questions and setting short- and long-term goals. You have to know exactly what information you need, or you won't be able to figure out how to get the data.
Find out what types of customers make a product successful, such as what drives customers to leave your brand. Then, collect as much data as you can about your current and past customers.
Finally, use this information to forecast which products will become popular with customers. You can also see when a customer is about to leave you, and so you can devise a strategy to entice them to come back to you.
Big Data poses greater challenges in terms of security and ethics issues
Big Data is not without its inadequacies. You're not wrong in assuming that there's something wrong with storing huge amounts of data about every listener. Big Data takes privacy issues to the next level. It would be catastrophic if a certain amount of Big Data data fell into the wrong hands.
Apple and Amazon, for example, both have the credit card information of some 400 million customers. What does that mean for hackers or data thieves?
Even assuming that we can trust these companies with our data (which is not necessarily a correct assumption), there are still plenty of cases where data is stolen. So companies must protect customer data in addition to protecting their internal data.
In 2012, Google got into trouble when it revealed that its Street View software was collecting data from free Wi-Fi networks. This is the flip side of Big Data: it is threatening our privacy.
Giant companies like Google, Amazon and Facebook can tap into unlimited user data if they want to. If this worries you, you can consider using other alternative services like DuckDuckGo – a search engine that does not store user data.
Big Data will make products "smarter"
So where is the future for Big Data? Some have downplayed its impact on the consumer market.
For example, we will see a gradual shift from active data to passive data .
Today, much of their data is actively generated . That means we use the internet every day with laptops, smartphones and we create the data ourselves.
But more and more data will be generated passively . We will have cars, TVs and many other devices that will connect to the internet and will monitor our every move. These technologies will generate the data themselves.
This may sound like an invasion of privacy. However, it also means that technology will adapt itself to each of our specific behaviors.
In the future, technology will be able to use data more efficiently. Ipod designer Tony Fadel developed a program for his new company, Nest, for example.
One of Nest's products is a thermostat. This product will collect the user's data to automatically adjust the temperature according to their wishes.
You won't have to program the Nest after it collects your data – it will use the data itself to do so. For example, it might know that you prefer a cooler living room during the day and a warmer bedroom at night. The more you use it, the better it will understand and meet your needs.
The data is collected in an orderly and networked manner. So you will be able to adjust the temperature with your smartphone and review past data. It is clear that Big Data is having a great influence on technology, the environment and our lives. Will you become a part of it?
Final Summary
The main message of the book:
Big Data requires a new way of thinking, new tools, a new approach to data analysis in general. However, if you can manage unstructured data, the benefits it brings are immense. Big Data will play an increasingly important role in our future. So don't miss out! Practical advice: Experiment! Feeling still drowsy about using Big Data? Actually you don't have to do it directly. Ask a Big Data company like Kaggle to analyze some of the data for you. You will understand some issues and know if you are ready for Big Data or not. Don't rush – slowly but surely and make sure you are prepared for it.
